Tunneling Interlayer for Efficient Transport of Charges in Metal Oxide

Subscriber access provided by NEW YORK UNIV

Article

Tunneling Interlayer for Efficient Transport of Charges in Metal Oxide Electrodes Jiazang Chen, Liping Zhang, Zhenhui Lam, Hua Bing Tao, Zhiping Zeng, Hong Bin Yang, Jianqiang Luo, Lin Ma, Bo Li, Jianfeng Zheng, Suping Jia, Zhijian Wang, Zhenping Zhu, and Bin Liu J. Am. Chem. Soc., Just Accepted Manuscript • DOI: 10.1021/jacs.5b13464 • Publication Date (Web): 11 Feb 2016 Downloaded from http://pubs.acs.org on February 19, 2016

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Journal of the American Chemical Society is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Journal of the American Chemical Society

Tunneling Interlayer for Efficient Transport of Charges in Metal Oxide Electrodes Jiazang Chen,

† ,‡

†

†

†

†

Liping Zhang, Zhenhui Lam, Hua Bing Tao, Zhiping Zeng, Hong Bin Yang,

†

† † ‡ ‡ ‡ ‡ Jianqiang Luo, Lin Ma, Bo Li,§ Jianfeng Zheng, Suping Jia, Zhijian Wang, Zhenping Zhu,*, and

Bin Liu*,† †

School of Chemical & Biomedical Engineering, Nanyang Technological University, 62 Nanyang Drive, Singapore 637459, Singapore; ‡ State Key Laboratory of Coal Conversion, Institute of Coal Chemistry, Chinese Academy of Sciences, Taiyuan 030001, China; §Lanzhou Petrochemical Research Center of Petrochemical Research Institute, PetroChina, Lanzhou 730060, China

*

To

whom

correspondence

should

be

addressed.

E-mail:

[email protected]

(Z.Z);

and

[email protected] (B.L.).

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 19

ABSTRACT: Due to the limited electronic conductivity, the application of many metal oxides that may have attractive (photo)-electrochemical properties has been limited. Regarding these issues, incorporating low dimensional conducting scaffolds into the electrodes or supporting the metal oxides onto the conducting networks are common approaches. However, some key electronic processes like interfacial charge transfer are far from being consciously concerned. Here we use carbon-TiO2 contact as a model system to demonstrate the electronic processes occurred at the metal-semiconductor interface. To minimize the energy dissipation for fast transfer of electrons from semiconductor to carbon scaffolds, facilitating electron tunneling while avoiding high energy-consuming thermionic emission is desired, according to our theoretical simulation of the voltammetric behaviors. To validate this, we manage to sandwich ultrathin TiO2 interlayers with heavy electronic doping between the carbon conductors and dopant-free TiO2. The radially graded distribution of the electronic doping along the cross-sectional direction of carbon conductor realized by immobilizing the dopant species on the carbon surface can minimize the energy consumption for contacts to both the carbon and the dopant-free TiO2. Our strategy provides an important requirement for metal oxide electrode design.

Keywords: electron tunneling, thermionic emission, charge transport, change transfer, metal oxide electrodes


2

Page 3 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1. Introduction Due to the limited electronic conductivity, the intrinsic electrochemical characteristics of many metal oxides that may have promising electrocatalytic activity for interfacial reactions are severely restricted.15

The same phenomenon is also common in photoelectrochemical (PEC) cells, in which the collection of

charges is less efficient than the ability to generate upon photo-induced processes, because of the limited charge transport in the semiconductor electrodes.6 Thus far, many approaches, including formation of aligned arrays,7,8 modulation of the intraband electronic states of the semiconductor,9-14 and the most exemplificative strategy, adoption of heterogeneous three-dimensional conducting networks (carbon nanostructures such as carbon nanotubes and graphene sheets are the most common) as special charge transport channels to incorporate into the semiconductor electrodes and/or to support the metal oxide nanostructures have been proposed.3,4,15-26 Despite intensive efforts, facilitating the transport of charges in the semiconductor electrodes is still far from being satisfactory. Transport of charges in the semiconductor electrodes based on these elaborately designed electrodes is only marginally faster than that in nanoparticulate films,27-29 due to the larger internal energy dissipation arising from the homogeneous/heterogeneous boundary charge transfer compared with that in the case of pure nanoparticles.27-30 Although charge transfer within the semiconductor electrodes plays a predominant role in limiting transport, the relationship between the electrode configuration and transport of charges, which has been expected to guide the designing of efficient (photo)-electrochemical devices, is undesirably obscured.31 Transport of charges in semiconductor electrodes is the most concerned issue. However, the designing of electrode configuration is in some extent intuitive, as the information on the electronic processes occurring in the semiconductor electrodes is usually hidden from the tangible investigations (e.g.: electron microscopy).31,32 It is believed that in carbon nanostructures incorporated TiO2 electrodes, the electrons can directly transfer to the conducting substrate via the carbon conductors, without encountering the numerous grain boundaries during the transport of electrons through the mesoporous films.23,25,33-37 However, the behaviors of the charge transfer through the heterogeneous interface from ACS Paragon Plus Environment

3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 19

TiO2 to the carbon network have hardly been investigated,30 and the potential-dependent energy consumption for the charges across the barriers following thermionic emission in some PEC and electrochemical cells is too large to be tolerable.30,38-40 Electron tunneling offers a potential solution by reducing the internal energy dissipation for charge transfer through the semiconductor-metal (S-M) interface.41-44 Here, we present an analysis to estimate the charge transfer dependent energy dissipation and propose a strategy to facilitate charge transfer by electron tunneling. To generalize our strategy, we adopt carbon nanotubes (CNTs) incorporated TiO2 based PEC devices to demonstrate the energy saving by electron tunneling, because the carbon-TiO2 contact is very common in many applications not limited to composite electrodes, but also for the carbon (graphite or glassy carbon) supported electrocatalysts/photocatalysts. Furthermore, this type of cells possesses substantially high current densities (e.g.: 5-20 mA/cm2), making them representative for the most of the (photo)-electrochemical systems. 2. Results and discussion 2.1. Development of Electron Tunneling Model for CNTs-TiO2 Interfacial Charge Transfer. Transfer of electrons from TiO2 into the carbon nanostructures is not only the prerequisite to play the role of charge transport channels in semiconductor electrodes, but also a basic requirement for the efficient charge separation for carbon composited TiO2 electrodes/(photo)-catalysts. According to contact theory, transfer of electrons through the S-M interface can be realized by two major approaches, thermionic emission and electron tunneling.41 Under forward bias, the manner of charge transfer on the S-M interface mainly depends on the height ( φ B ) and the width (w) of the potential barriers (Figure 1). To remain the intrinsic properties and/or reduce the charge loss by interfacial recombination, the TiO2 used for fabrication of (photo)-electrochemical cells (e.g.: dye-/quantum dots-sensitized solar cells, DSSCs/QDSSCs) is almost dopant-free. Thus, it would be appropriate to apply thermionic emission to describe the transfer of charges from the semiconductor to the metals (e.g.: carbon).30,39 However, charge transfer by thermionic emission will consume large amount of energy if the Schottky barrier height is larger than 0.5 eV (Figure S1), which will result undesirably in a lowering of energy output.38ACS Paragon Plus Environment

4

Page 5 of 19 40

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


For the case of carbon and TiO2, the height of Schottky barrier reaches as high as 0.75 eV (by

adopting 4.95 eV as the work function for carbon nanotubes and -4.2 eV for the conduction band level for TiO2).45 Thus, it is not surprising that the photocurrent-voltage (J-V) characteristics of the devices based on carbon nanostructures composited TiO2 electrodes offset the normal photovoltaic behaviors of semiconductor-liquid junction cells.18,20-22,25,37,46-50 Reducing the width of the potential barriers to increase the probability of electron tunneling seems to be a rational approach to facilitate the charge transfer through the S-M interface and to reduce the internal energy dissipation (Figure S2, Figure S3, and Figure S4). Generally, reducing the width of the potential barriers can be simply realized by increasing the density of electronic doping in the semiconductor (Figure 1d and Figure S4).51 However, increasing the density of electronic doping will simultaneously increase the probability of trapping/detrapping events during the operation of the cells and aggravate the charge loss at semiconductor-electrolyte interface (SEI).52 To fully exert the function of the carbon conductor to facilitate the charge collection while avoiding severe charge recombination at SEI, it is rational to selectively dope the semiconductor to electrically bridge the S-M interface (Figure 1e). In this work, inspired by the simulated J-V behaviors (Figure 2a) with consideration of electron tunneling, we adopt heavily electronic doped ultrathin TiO2 layers as the intermediate to connect the doping-free TiO2 nanoparticles and the CNTs (Figure 1e, Figure 2b). To demonstrate the intrinsic superiority of the intermediate bridging, electronic process manner and PEC behaviors occurring in the CNT-TiO2 electrodes were investigated. Figure 2b shows the schematic diagram of the electrode configuration as we proposed. The intermediate ultrathin electronically doped TiO2 layers are sandwiched between the carbon conductors and the doping-free TiO2 building blocks in the electrodes. One important feature for the intermediate layers is the density of electronic doping, which is graded distributed along the radial direction of the CNTs. To facilitate the charge transfer by electron tunneling, the density of electronic doping near the SM interface is kept high; while it becomes lower as walking away from the CNTs. On the boundary to the doping-free TiO2, the inter-grain difference in electronic structures disappears. In this situation, the ACS Paragon Plus Environment

5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 19

electrons in the TiO2 nanostructures, which are photogenerated by light excitation or injected from other excited light absorbers, can be collected by the tubular carbon nanostructures by passing through the heavily electronic doped TiO2 ultrathin layers, without severely consuming the energy. Besides, in this type of electrode configuration, the introduction of electronic doping will not aggravate the charge loss by interfacial recombination, owing to the fact that the highly doped TiO2 is concentrated in the vicinity of the interface to CNTs and has no chance to access the electron acceptors in the electrolyte. 2.2. Growth of Graded-Distributed Niobium-Doped TiO2 Ultrathin Layer on Carbon Nanotubes. To validate the theoretical prediction, it is required to prepare electronically doped TiO2 ultrathin layer coated CNTs. Niobium is a common foreign element to realize electronic doping for TiO2 with a wide range of dopant concentrations.53 To achieve graded distribution of the niobium dopants along the radial direction of CNTs, we firstly immobilize the guest metal molecules (ions) on the surface of the carbon nanostructures by chelating the surface functional groups like carboxylic groups with their empty dorbitals (Figure S5).54 After that, ethylenediamine is employed to connect the dopant precursor with the host titanium-containing molecules through its electrophilic group in the tips of the molecules (Figure S5, Figure S6). Finally, the ultrathin TiO2 layer with radially graded distribution of niobium dopants (denoted CNT|Nb-TiO2(t)) can be grown on CNTs by dehydrating the precursor-coated CNTs (Figure 3, Figure S5, Figure S6, and Figure S7). To demonstrate the effect of electronic doping, TiO2 ultrathin layer coated CNTs without niobium dopant (denoted CNT|TiO2(t)) was also prepared by a similar method without introducing niobium precursor (Figure S8). Figure 3a-c present the scanning electron microscopy (SEM) images of the CNTs before and after modification of ultrathin (niobium-doped) TiO2 layers. The increase in the diameter and the surface roughness suggest the successful coating of TiO2 layers on the CNTs. By introduction of niobium precursor into the reaction system, it can be seen that the dopant is of graded distribution along the radial direction of CNTs (Figure S9) and does not alter the thickness of TiO2 coating layers (Figure S7). X-ray diffraction (XRD) patterns show that both TiO2 ultrathin coatings with and without niobium doping are of anatase phase (Figure 3d). The absence of niobium oxide and other niobium-containing species in the ACS Paragon Plus Environment

6

Page 7 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


XRD pattern suggests that all niobium are doped into the TiO2 layers. The significant broadening of the (101) diffraction peak for the CNT|Nb-TiO2(t) sample indicates distortion of the TiO2 lattice, resulting most likely from the doping by niobium (inset of Figure 3d).53 Raman spectroscopy provides a detailed structural probe of local coordination environments.55 The broadening of all Raman peaks and the blue shift of the characteristic Eg(1) mode as well as the red shift of Eg(3) mode suggest that the electronic structures have been modified by introduction of niobium into the TiO2 (Figure 3e). Besides, the appearance of the Nb 3d signal and broadening of Ti 2p peaks in the X-ray photoelectron spectroscopy (XPS) spectra further confirms the successful niobium doping in the CNT|Nb-TiO2(t) sample (Figure S10). 2.3. Energy Saving by Electron Tunneling. As studied previously,19,33 the introduction of conducting scaffold into mesoporous TiO2 electrodes can improve the charge collection by reducing the boundary scattering of electrons during the charge transport processes. The increase in the short-circuit photocurrent density as shown in the photovoltaic characteristics reveals that the introduction of CNTs into mesoporous TiO2 electrodes can indeed improve the electron transport (Figure 4). However, due to different charge transfer behaviors across the S-M interface, the increment of photocurrent for individual cell is rather different. The reference cell without CNTs exhibits the lowest photocurrent density (JSC, 10.18 mA/cm2) and energy conversion efficiency (η, 5.31 %, Figure 4a). Following an ascending order in performance is the cell using CNTs incorporated TiO2 (CNT/TiO2) electrodes, which exhibits 13.43 mA/cm2 and 6.55 % for JSC and η, respectively. The cell with CNT|TiO2(t) incorporated TiO2 electrodes exhibits a continuously increased photocurrent density (JSC = 14.55 mA/cm2) and energy conversion efficiency (η = 6.94 %), due likely to the suppression of interfacial charge leakage via the carbon nanostructures.30 The cells with CNT|Nb-TiO2(t) incorporated TiO2 electrodes exhibits the best performance, showing 8.69 % of energy conversion efficiency and 16.92 mA/cm2 of short-circuit photocurrent density. Among the photovoltaic parameters, it is clear that JSC is very sensitive to the electron transport processes, which plays the predominant role on the variation of η for all of the cells.


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 19

The improvement in charge collection for the CNTs-containing TiO2 electrodes is strongly dependent on the transfer behaviors of the photogenerated electrons from the semiconductor to the carbon conducting scaffolds. This is to say that the energy dissipation for the transfer of charge through the S-M interface eventually determines the role of the charge transport channels. For the CNT/TiO2 electrode, the situation is very similar to the Schottky contact of doping-free semiconductor to metal, in which thermionic emission is the predominant manner for transfer of charges. As the potential barriers for CNT-TiO2 contact reaches as high as 0.75 eV, the energy consumption for the charge transfer across the TiO2-CNT interface is very high (Figure S1). As a result, only certain percentage of electrons can overcome the barrier and transfer into the carbon conductor. The situation becomes even worse in the region where the band bending is large (Figure 4b, c).30 Therefore, the function of charge transport channels will be restricted and improvement in the charge collection will be limited in CNT/TiO2 electrode (Figure 4a, d). Introducing electronically doped TiO2 ultrathin layers into the CNTs incorporated electrodes (CNT|Nb-TiO2(t)/TiO2) greatly improves the charge collection and charge diffusion coefficient (Dn). It is well known that the probability of electron tunneling from semiconductor to metal increases with increasing in density of electronic doping in the semiconductor phase (Figure S4), which in-turn lowers the energy consumption for charge transfer (Figure S2). Here in our case, the average electronic doping density of the TiO2 in the CNT|Nb-TiO2(t) sample reaches as high as 1.79 × 1021 cm-3 (Figure S11), which is much higher than that of common doping-free TiO2 nanoparticles (7.15 × 1018 cm-3). Thus, in our CNT|Nb-TiO2(t)/TiO2 electrodes, the carbon networks could fully exert the function of charge transport channels by reducing the energy dissipation for charge transfer across the TiO2-CNT interfaces via electron tunneling through highly electronically doped intermediate layers. Therefore, electrons will take the energetically favorable pathways to reach the substrate by eliminating the multiple TiO2 particle-particle boundary scatterings. Since the energy dissipation for charge transfer through the S-M interface is very low, the improvement in charge transport is still evident even in the case of low


8

Page 9 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


photocurrent densities, in which the band bending is very high and the thermionic emission is almost forbidden (Figure 4b, c, d). It is noteworthy that the introduction of CNT|TiO2(t) into the TiO2 electrodes not only improves the charge collection and JSC, but also increases the charge diffusion coefficient (Figure 4b, c), as compared with the case of CNT/TiO2 electrode, especially in the region of low photocurrent densities, in which the thermionic emission is insignificant. It is expected that coating CNTs with ultrathin TiO2 layers can improve the charge collection by effectively suppressing interfacial charge leakage via the naked part of carbon.56 The increase in charge diffusion coefficient should be ascribed to electron tunneling due to the following reason: generally, in nanostructured semiconductor electrodes, the major potential drop can be confined in the first few layers of the nanoparticles in contact to the metal-like conductor (Figure S12).57 In our case, the thickness of the ultrathin TiO2 coating in CNT|TiO2(t) is only ~ 4 nm (Figure S7), which is similar to the critical thickness to tunnel a considerable current.42-44 As a result, transfer of charges through the S-M interface by means of electron tunneling should occur favorably, which is helpful to facilitate charge transport as well as increase the charge diffusion coefficient (Figure 4b, c, d). To further verify the validity of the merits of electron tunneling to charge collection at low current conditions, we examined the photo-induced voltammetric behaviors of naked TiO2 electrodes in a trielectrode water splitting system (Figure 5). As the porous electrode is unsensitized, it can only generate a small photocurrent by absorbing the ultraviolet photons in the solar spectrum. The PEC results once again confirm that the heavily electronically doped TiO2 ultrathin layers can offer efficient charge transfer through the S-M interface by means of electron tunneling and improve the charge collection (Figure 5). Since the photocurrent is generally low, charge transfer through the TiO2-CNT by thermionic emission seems impossible.30 As a result, the carbon conductors hardly have any benefit to the charge transport or charge collection. The slightly decrease in photocurrent for the CNT/TiO2 electrode should be ascribed to the leakage of the photogenerated electrons via the naked carbon nanostructures (Figure 5).56


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 19

To further popularize the resultant energy savings of electron tunneling in PEC applications, we demonstrate the virtue of our electrode configuration in visible light driven water splitting. Similar to the situation of dye-sensitized photovoltaic devices, the cell with CNT|Nb-TiO2(t) incorporated electrode exhibits the best performance for hydrogen evolution, showing the highest photocurrent over the measured range and a plateau current density of ~4.6 mA/cm2 over the potential window of -1.0 – -0.3 V vs. SCE (Figure 6). Following a descending order in performance is the cell incorporated with CNT|TiO2(t), which exhibits a plateau current density of 3.7 mA/cm2. Cell with plain TiO2 electrodes exhibits the lowest plateau current density of 2.4 mA/cm2. For the cell using CNT/TiO2 electrodes, the plateau current density is 3.1 mA/cm2. Compared with the case of plain TiO2 electrode (Figure 5), the higher photocurrent for the CNT/TiO2 cells suggests that certain percentage of photogenerated electrons can overcome the potential barrier following the thermionic emission at such current density, which is in good agreement with the relationship of Dn-JSC for the dye-sensitized solar cells as shown in Figure 4c. 3. Conclusion In this work, we have studied the role of tunneling interlayers in the carbon conductors incorporated semiconductor electrodes as an effective way to save the energy from excessive internal dissipation, which many metal oxides encounter during transfer of charges over the potential barriers by thermionic emission. We have employed CNT-TiO2 contact as the model systems to demonstrate the charge transfer behaviors and related energy consumption. On the basis of theoretical simulation, we found that the electron tunneling could effectively save the energy dissipation for charge transfer through the metalsemiconductor interface and fully exert the role of the charge transport channels, which is in good agreement with the charge diffusion parameters. By application of tunneling interlayers into the CNTs incorporated TiO2 electrodes, the facilitated S-M interfacial transfer of charges can improve charge transport and collection in every circumstance of high (photovoltaic devices), mediate (visible light induced water splitting), and low (UV induced water splitting) current densities. We believe that our work can also prompt the sequential development of perovskite solar cells, energy storage devices, oxygen evolution reaction and other relevant applications, in which metal oxides are employed. ACS Paragon Plus Environment

10

Page 11 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Acknowledgments. The authors acknowledge A*Star (M4070178.120), Nanyang Technological University (M4080977.120), State Key Laboratory of Coal Conversion (J15-16-913), and Ministry of Education of Singapore (M4011021.120) for the financial support. J. Chen thanks Dr. Yinghua Xu (Zhejiang Univ. Technol.) and Dr. Tiancheng Xu (HzCell Electrochem. Corp.) for the help on discussion of photoelectrochemical results. Acknowledgements are also given to Dr. Kedar Hippalgaonkar and Dr. Chiam Sing Yang (Institute of Materials Research and Engineering, A*Star) for the XPS measurements. Supporting Information Available. Simulation and Experimental details, Figure S1-S13. This material is available free of charge via the Internet at http://pubs.acs.org. References and Notes (1) (2) (3)

Chen, C. T.; Cahan, B. D. J. Electrochem. Soc. 1982, 129, 17. Lyons, M. E. G.; Brandon, M. P. J. Electroanal. Chem. 2009, 631, 62. Jin, H.; Wang, J.; Su, D.; Wei, Z.; Pang, Z.; Wang, Y. J. Am. Chem. Soc. 2015, 137,

2688. (4) Liang, H.; Meng, F.; Cabán-Acevedo, M.; Li, L.; Forticaux, A.; Xiu, L.; Wang, Z.; Jin, S. Nano Lett. 2015, 15, 1421. (5) Morales-Guio, C. G.; Mayer, M. T.; Yella, A.; Tilley, S. D.; Grätzel, M.; Hu, X. J. Am. Chem. Soc. 2015, 137, 9927. (6) Hagfeldt, A.; Boschloo, G.; Sun, L. C.; Kloo, L.; Pettersson, H. Chem. Rev. 2010, 110, 6595. (7) Liu, B.; Aydil, E. S. J. Am. Chem. Soc. 2009, 131, 3985. (8) Qian, F.; Wang, G.; Li, Y. Nano Lett. 2010, 10, 4686. (9) Yang, X.; Wolcott, A.; Wang, G.; Sobo, A.; Fitzmorris, R. C.; Qian, F.; Zhang, J. Z.; Li, Y. Nano Lett. 2009, 9, 2331. (10) Ling, Y.; Wang, G.; Wheeler, D. A.; Zhang, J. Z.; Li, Y. Nano Lett. 2011, 11, 2119. (11) Wang, G.; Wang, H.; Ling, Y.; Tang, Y.; Yang, X.; Fitzmorris, R. C.; Wang, C.; Zhang, J. Z.; Li, Y. Nano Lett. 2011, 11, 3026. (12) Wang, Q.; Zhang, Z.; Zakeeruddin, S. M.; Gratzel, M. J. Phys. Chem. C 2008, 112, 7084. (13) Chandiran, A. K.; Sauvage, F.; Casas-Cabanas, M.; Comte, P.; Zakeeruddin, S. M.; Graetzel, M. J. Phys. Chem. C 2010, 114, 15849. (14) Lu, X. J.; Mou, X. L.; Wu, J. J.; Zhang, D. W.; Zhang, L. L.; Huang, F. Q.; Xu, F. F.; Huang, S. M. Adv. Func. Mater. 2010, 20, 509. (15) Hu, Y.; Su, Y.; Huang, H.; Qian, Q.; Guan, Z.; Feng, J.; Li, Z.; Zou, Z. ChemCatChem 2015, 7, 2979. (16) Sun, J.; Memon, M. A.; Bai, W.; Xiao, L.; Zhang, B.; Jin, Y.; Huang, Y.; Geng, J. Adv. Func. Mater. 2015, 25, 4334. (17) Tu, X.; Luo, S.; Chen, G.; Li, J. Chem. Eur. J. 2012, 18, 14359. (18) Kongkanand, A.; Martínez Domínguez, R.; Kamat, P. V. Nano Lett. 2007, 7, 676. (19) Kamat, P. V. J. Phys. Chem. Lett. 2011, 2, 242. 11 ACS Paragon Plus Environment


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

(20)

Page 12 of 19

Chen, T.; Qiu, L.; Cai, Z.; Gong, F.; Yang, Z.; Wang, Z.; Peng, H. Nano Lett. 2012, 12,

2568. (21) Tang, Y. B.; Lee, C. S.; Xu, J.; Liu, Z. T.; Chen, Z. H.; He, Z. B.; Cao, Y. L.; Yuan, G. D.; Song, H. S.; Chen, L. M.; Luo, L. B.; Cheng, H. M.; Zhang, W. J.; Bello, I.; Lee, S. T. ACS Nano 2010, 4, 3482. (22) Yang, N. L.; Zhai, J.; Wang, D.; Chen, Y. S.; Jiang, L. ACS Nano 2010, 4, 887. (23) Kamat, P. V.; Tvrdy, K.; Baker, D. R.; Radich, J. G. Chem. Rev. 2010, 110, 6664. (24) Hayashi, H.; Lightcap, I. V.; Tsujimoto, M.; Takano, M.; Umeyama, T.; Kamat, P. V.; Imahori, H. J. Am. Chem. Soc. 2011, 133, 7684. (25) Dang, X.; Yi, H.; Ham, M.-H.; Qi, J.; Yun, D. S.; Ladewski, R.; Strano, M. S.; Hammond, P. T.; Belcher, A. M. Nat. Nano. 2011, 6, 377. (26) Chen, R.; Wang, H.-Y.; Miao, J.; Yang, H.; Liu, B. Nano Energy 2015, 11, 333. (27) Zhu, K.; Neale, N. R.; Miedaner, A.; Frank, A. J. Nano Lett. 2007, 7, 69. (28) Zhu, K.; Vinzant, T. B.; Neale, N. R.; Frank, A. J. Nano Lett. 2007, 7, 3739. (29) Jennings, J. R.; Ghicov, A.; Peter, L. M.; Schmuki, P.; Walker, A. B. J. Am. Chem. Soc. 2008, 130, 13364. (30) Chen, J.; Li, B.; Zheng, J.; Zhao, J.; Zhu, Z. J. Phys. Chem. C 2012, 116, 14848. (31) Villanueva-Cab, J.; Jang, S.-R.; Halverson, A. F.; Zhu, K.; Frank, A. J. Nano Lett. 2014, 14, 2305. (32) Chen, J.; Yang, H. B.; Miao, J.; Wang, H.-Y.; Liu, B. J. Am. Chem. Soc. 2014, 136, 15310. (33) Roy-Mayhew, J. D.; Aksay, I. A. Chem. Rev. 2014, 114, 6323. (34) Jariwala, D.; Sangwan, V. K.; Lauhon, L. J.; Marks, T. J.; Hersam, M. C. Chem. Soc. Rev. 2013, 42, 2824. (35) Huang, X.; Qi, X. Y.; Boey, F.; Zhang, H. Chem. Soc. Rev. 2012, 41, 666. (36) Costa, R. D.; Lodermeyer, F.; Casillas, R.; Guldi, D. M. Energy Environ. Sci. 2014, 7, 1281. (37) Ng, Y. H.; Lightcap, I. V.; Goodwin, K.; Matsumura, M.; Kamat, P. V. J. Phys. Chem. Lett. 2010, 1, 2222. (38) Brown, P.; Takechi, K.; Kamat, P. V. J. Phys. Chem. C 2008, 112, 4776. (39) Ni, M.; Leung, M. K. H.; Leung, D. Y. C.; Sumathy, K. Solar Energy Mater. Solar Cells 2006, 90, 2000. (40) Kron, G.; Egerter, T.; Werner, J. H.; Rau, U. J. Phys. Chem. B 2003, 107, 3556. (41) Sze, S. M.; Ng, K. K. Physics of Semiconductor Devices; 3rd ed.; John Wiley $ Sons: Hoboken, 2007. (42) Viswanathan, V.; Pickrahn, K. L.; Luntz, A. C.; Bent, S. F.; Nørskov, J. K. Nano Lett. 2014, 14, 5853. (43) Kim, J.; Kim, B.-K.; Cho, S. K.; Bard, A. J. J. Am. Chem. Soc. 2014, 136, 8173. (44) Hill, C. M.; Kim, J.; Bard, A. J. J. Am. Chem. Soc. 2015, 137, 11321. (45) Shiraishi, M.; Ata, M. Carbon 2001, 39, 1913. (46) Guo, W.; Xu, C.; Wang, X.; Wang, S.; Pan, C.; Lin, C.; Wang, Z. L. J. Am. Chem. Soc. 2012, 134, 4437. (47) Manga, K. K.; Zhou, Y.; Yan, Y. L.; Loh, K. P. Adv. Funct. Mater. 2009, 19, 3638. (48) Kusumawati, Y.; Martoprawiro, M. A.; Pauporté, T. J. Phys. Chem. C 2014, 118, 9974. (49) Brown, P.; Takechi, K.; Kamat, P. V. J. Phys. Chem. C 2008, 112, 4776. (50) Xiang, Z. H.; Zhou, X.; Wan, G.; Zhang, G. X.; Cao, D. P. ACS Sustain. Chem. Eng. 2014, 2, 1234. (51) Neamen, D. An Introduction to Semiconductor Devices; McGraw-Hill, Inc., 2006. (52) Peter, L. M. Chem. Rev. 1990, 90, 753. (53) Wang, H.-Y.; Chen, J.; Xiao, F.-X.; Zheng, J.; Liu, B. J. Mater. Chem. A 2016, 10.1039/c5ta08202a. 12 ACS Paragon Plus Environment

Page 13 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(54) Livage, J.; Henry, M.; Sanchez, C. Prog. Solid State Chem. 1988, 18, 259. (55) Swamy, V.; Kuznetsov, A.; Dubrovinsky, L. S.; Caruso, R. A.; Shchukin, D. G.; Muddle, B. C. Phys. Rev. B 2005, 71, 184302. (56) Kyaw, A. K. K.; Tantang, H.; Wu, T.; Ke, L.; Peh, C.; Huang, Z. H.; Zeng, X. T.; Demir, H. V.; Zhang, Q.; Sun, X. W. Appl. Phys. Lett. 2011, 99, 021107. (57) Ruhle, S.; Cahen, D. J. Phys. Chem. B 2004, 108, 17946.


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 19

Figures and Captions

Figure 1. Energy band diagrams for isolated metal and semiconductor (a) and the formed metalsemiconductor contact (b-e). It can be seen that the band in the semiconductor side of the formed contact is bent upward (b-e). The potential barrier height ( ϕ B ) of the metal-semiconductor contact, can be given by: ϕ B = Wm − χ , where Wm is work function of the metal, χ is the electron affinity of the semiconductor. On the semiconductor side, the built-in potential barrier (qVbi) can be obtained by: qVbi = ϕ B − ϕ n − qVD , in which VD is the bias potential applied to the contact. In the case of doping-free semiconductor, the barrier for transfer of electrons from semiconductor to the metal by thermionic emission becomes smaller as the forward bias increases (b, c). By electronic doping of the semiconductor, the width of the space charge layer becomes narrower and electron tunneling emerges to be the predominant charge transfer behaviors (d). Efficient electron tunneling can also be realized by selectively doping the semiconductor, in which the semiconductor is only electronically doped near the S-M interface, while the bulk remains dopant-free (e).


14

Page 15 of 19

20

a

15 2

J (mA / cm )

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


10 TiO2

5

CNT-TiO2 (TE) CNT-TiO2 (ET)

0 0.0

0.2

0.4

0.6

0.8

E (V)

Figure 2. Features of the composite electrodes with efficient electron tunneling through the TiO2-carbon interface. (a) Effect of interfacial charge transfer behaviors on the J-V features of the photovoltaic cells based on CNTs incorporated TiO2 electrodes. The simulated J-V curves show that the charge transfer through the CNT-TiO2 interface by electron tunneling (ET) can effectively reduce the internal energy dissipation, as compared with the case of thermionic emission (TE). The parameters adopted for the simulation are: ideality factor, m = 1.8; saturation current, J0 = 1× 10-5 mA/cm2; series resistance, Rs = 10 ohm; height of carbon-TiO2 barrier, ϕ B = 0.75 eV; height of FTO-TiO2 barrier, ϕ B = 0.2 eV; the density of electronic doping of TiO2 for the case of electron tunneling is 1021 cm-3; the short-circuit photocurrent densities, JSC = 10.00 mA/cm2 for the plain cells, and increased to 14, and 18 mA/cm2 for the cells with CNTs incorporated TiO2 electrodes to show the benefit of charge collection improvement. ACS Paragon Plus Environment

15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 19

(b) Schematic illustration of hierarchical allocations of electronic doping in the CNTs incorporated TiO2 electrodes. The small sized TiO2 contact to the CNTs is heavily electronically doped by foreign elements like niobium.

d

e

Eg(1)

CNT | Nb-TiO2 (t ) CNT | TiO2 (t )

24

25

26

27

Eg(3)

B1g(1) A1g+B1g(2)

20 30 40 50 60 70 80 2 Theta (Cu Kα)

120

160 300 400 500 -1 Raman shift (cm )

600

700

Figure 3. Features of the ultrathin TiO2 coating layers on the CNTs. SEM images for the CNT-COOH before (a) and after coating with ultrathin TiO2 (b) and niobium-doped TiO2 (c) layers. XRD patterns (d) and Raman spectra (e) for CNT|TiO2(t) and CNT|Nb-TiO2(t) samples.


16

Page 17 of 19

20

10

a

6

Dn (10 cm / s)

4

2

2

J (mA / cm )

b

8

15

-5

10 TiO2 CNT/TiO2

5

2

CNT | TiO2(t )/TiO2 CNT | Nb-TiO2(t )/TiO2

0 0.0

-0.2

-0.4

-0.6

-0.8

0.0

0.1

0.2

0.3

0.4

0.5

0.6

E (vs VOC, V)

E (V)

c 10

-4

10

-5

10

-6

2

Dn (cm / s)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


10

-2

10

-1

10 2 JSC (mA / cm )

0

10

1

Figure 4. Photovoltaic characteristics of the cells based on CNTs-TiO2 electrodes. (a) Photocurrentvoltage characteristics of the cells with various CNT-TiO2 electrodes. (b) Diffusion coefficients of photogenerated electrons in the semiconductor electrodes of the cells under bias versus the open-circuit potentials. (c) Diffusion coefficients of the photogenerated electrons in semiconductor electrodes under various illumination intensities. (d) Schematic diagrams for the electronic processes occurring in CNTs incorporated TiO2 electrodes with various situations of CNT-TiO2 contact. The dashed arrow shown in CNT/TiO2 electrode (upper left) indicates the transfer of charges from TiO2 to CNTs by thermionic emission is not always available. By placing the ultrathin TiO2 layers between the CNTs and dopant-free TiO2, charge transfer through the CNT-TiO2 interface becomes considerable as electron tunneling 17 ACS Paragon Plus Environment


emerges (thin arrow in the bottom left panel). In the CNT|Nb-TiO2(t) incorporated TiO2 electrode, CNTs fully exert the role of charge transport channels as the electron tunneling for the charge transfer through the CNTs-TiO2 interface is very obvious.

40

40

a

b

30 30 2

2

J (µA / cm )

20

J (µA / cm )

10 TiO2 0

CNT/TiO2

20

10

CNT | TiO2(t )/TiO2 -10 -20 0.1

CNT | Nb-TiO2(t )/TiO2 0.0

0

-0.1 -0.2 -0.3 -0.4 -0.5

0

20

40

E (vs SCE, V)

60

80

Time (s)

Figure 5. Voltammetric behaviors (a) and photocurrent responses (b) of the PEC cells under AM 1.5G illumination.

8 CNT | Nb-TiO2(t )/TiO2 CNT | TiO2(t )/TiO2 6

CNT/TiO2 TiO2

2

J (mA / cm )

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 19

4

2

0 -1.5

-1.0

-0.5

0.0

E (vs. SCE, V)

Figure 6. Voltammetric behaviors of the PEC cells based on CdS quantum dots-sensitized TiO2 electrodes under AM 1.5G illumination.


18

Page 19 of 19

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


TOC


19

Tunneling Interlayer for Efficient Transport of Charges in Metal Oxide

Recommend Documents